WatsonPaths: Scenario-Based Question Answering and Inference over Unstructured Information

Lally, Adam (Information Technology and Services) | Bagchi, Sugato (IBM Research) | Barborak, Michael A. (IBM T. J. Watson Research Center) | Buchanan, David W. (IBM T. J. Watson Research Center) | Chu-Carroll, Jennifer (IBM Research) | Ferrucci, David A. (Bridgewater) | Glass, Michael R. (IBM Research) | Kalyanpur, Aditya (IBM T. J. Watson Research Center) | Mueller, Erik T. (Capital One) | Murdock, J. William (IBM T. J. Watson Research Center) | Patwardhan, Siddharth (IBM T. J. Watson Research Center) | Prager, John M. (IBM T. J. Watson Research Center)

AI Magazine 

We present WatsonPaths, a novel system that can answer scenario-based questions. These include medical questions that present a patient summary and ask for the most likely diagnosis or most appropriate treatment. WatsonPaths builds on the IBM Watson question answering system. WatsonPaths breaks down the input scenario into individual pieces of information, asks relevant subquestions of Watson to conclude new information, and represents these results in a graphical model. Probabilistic inference is performed over the graph to conclude the answer. On a set of medical test preparation questions, WatsonPaths shows a significant improvement in accuracy over multiple baselines.

Duplicate Docs Excel Report

Title
None found

Similar Docs  Excel Report  more

TitleSimilaritySource
None found